Fair enough... I wasn't sure from your original post if you'd considered issues such as pressure and mass flow vs volumetric flow but it sounds like you have plenty of experience in this area.
One thing that does occur to me though is that there's nothing to stop you connecting two MFM's with different ranges in series. Then your software can choose each time whether to take the reading from the low-range MFM, or the high-range MFM if the low-range one is saturated. I think MFM's have a pretty high tolerance to overrange and I don't think a low-flow model will present much a bigger flow restriction than a high-flow one, but you'd need to check this. If you decide the piston is the way to go, have you considered using some kind of linear displacement transducer instead of image analysis? What kind of volume do you need to measure over what time scale? Is the process that evolves the hydrogen insensitive to pressure changes or what is the maximum back pressure you can accept? There must be lots of other possible ways to do it e.g. have the gas displace a liquid into a container on a balance, mass spectrometry... ;-) You could ask a calibration lab or a company such as Chell (http://www.chell.co.uk) about volume calibration equipment - or how about the manufacturer of your MFC's? (as a customer of MKS UK in the past I've found them very approachable when I had questions). Hope this helps, and hope that you also get some answers from IMAQ experts as you originally asked for!
